Tribological properties of nanoporous anodic aluminum oxide film

Tribological properties of nanoporous anodic aluminum oxide film

Surface & Coatings Technology 205 (2010) 1431–1437 Contents lists available at ScienceDirect Surface & Coatings Technology j o u r n a l h o m e p a...

2MB Sizes 2 Downloads 44 Views

Surface & Coatings Technology 205 (2010) 1431–1437

Contents lists available at ScienceDirect

Surface & Coatings Technology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s u r f c o a t

Tribological properties of nanoporous anodic aluminum oxide film Hyo-sang Kim a, Dae-hyun Kim a, Woo Lee b, Seong Jai Cho b, Jun-Hee Hahn b,⁎, Hyo-Sok Ahn a,⁎ a b

Graduate school of Nano IT Fusion Technology, Seoul National University of Technology, 172 Gongreung 2-dong, Nowon-gu, Seoul 139-743, South Korea Division of Industrial Metrology, Korea Reserach Institute of Standards and Science (KRISS), 267 Gajeong-ro, Yuseong-gu, Daejeon 305-340

a r t i c l e

i n f o

Available online 23 july 2010 Keywords: Anodic aluminum oxide Pore diameter Friction and wear Tribolayer Tribochemical reaction

a b s t r a c t Friction and wear properties of nanostructured anodic aluminum oxide (AAO)) films were studied in relation to contact load and pore size (pore diameter). Uniformly arrayed nanoporous aluminum oxide films (pores of 28 nm, 45 nm, 95 nm, and 200 nm diameter and 60–100 μm thick) were synthesized by anodization. Reciprocating wear tests using 1 mm diameter steel balls as counterpart were carried out for a wide range of load (from 1 mN to 1 N) at ambient environment. The friction coefficient reduced with the increase of load. The friction coefficient decreased by approximately 30% when the load increased by 3 orders of magnitude. The pore density marginally affected the frictional properties of AAO films. The influence of pore size on the friction coefficient was significant at relatively high loads (0.1 N and 1 N) whereas it was negligible at low loads (1 mN and 10 mN). The worn surface of AAO films tested at low loads did not experience tribochemical reaction and exhibited only mild plastic deformation. Dispersed thick smooth films were formed on the worn surface of all samples at relatively high loads whereas only extremely thin smooth film patches were rarely formed at low loads. These thick smooth films were generated by combined influence of tribochemical reaction at the contact interface and plastic deformation of compacted debris particles as evidenced by energy-dispersive spectroscopy analysis. We suggest that these thick films mainly contributed to the decrease of friction regardless of the pore size. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Nanostructured surface have drawn much attention for applications in various fields. One of the promising candidates is nanoporous anodic aluminum oxide (AAO) with self-organized hexagonal arrays of uniform parallel nanochannels. AAO films formed by electrochemical oxidation of aluminum have recently been proposed in medical, biological, electrical, optical and micromechanical fields as a template for the growth of low-dimensional nano-materials and as a threedimensional micro/nanostructure itself. These include nano-dots, nano-wires and nano-tubes, as well as photonic-crystal waveguide devices, nanocomposite structures, membranes for bio-separation, high-density magnetic memories, single-electron devices and biotechnological structures such as scaffold for tissue growth [1,2]. It is also recognized that AAO films are potentially advantageous for tribological application as the nanoporous structure can be utilized as a reservoir or template for solid lubricants [3–6] and nano-tubes, -rods or -fibers [7,8] to form self-lubricating structures. The researchers have investigated the tribological properties of their fabricated self-lubricating structure and demonstrated the improvement in friction and wear performance to some extent [3–8]. However, the tribological behavior of AAO film itself has rarely been

reported. We believe that a better understanding of AAO films is indeed a need to efficiently utilize them as the platform for facilitating optimal self-lubricating structures. The only available literature published dealt with the dependence of porosity on friction using an atomic force microscope [9]. As the probe used in the AFM had a radius of curvature at nanometer scale, the contact pressure applied in the contact became excessively high when compared with those experienced in MEMS devices, rendering their result less practical. In this study, we have examined the tribological behavior of AAO films employing more realistic contact conditions. The tribological properties of AAO films were investigated under conditions of the mesoscale contact (~nominal contact radius about from 2 to 21 μm). This involved the contact of a steel ball and local pressures/velocities comparable with that of conventional MEMS operating conditions. Nanoindentation tests were performed to measure hardness and elastic modulus of AAO films. Worn surfaces were analyzed using scanning electron microscopy (SEM) and chemical composition was identified by Energy-dispersive spectroscopy (EDS). Atomic force microscopy (AFM) was used to characterize surface topography. 2. Experimental details 2.1. Preparation of AAO films

⁎ Corresponding authors. E-mail addresses: [email protected] (J.-H. Hahn), [email protected] (H.-S. Ahn). 0257-8972/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.surfcoat.2010.07.056

Typical mild anodization processes is illustrated in Fig. 1. Selfordered anodic aluminum oxide films with desired interpore distance

1432

H. Kim et al. / Surface & Coatings Technology 205 (2010) 1431–1437 Table 1 Characteristics of nanoporous structured AAO films. Dp Dint ρ (pores/ (nm) (nm) cm2) 28 60 (MA) 45 105 (MA) 95 300 (HA) 200 500 (MA)

Porosity E (GPa) (%)

H (GPa)

Rq Thickness (nm) (μm)

2.9 × 1010

8.8

113.16 ± 1.59 5.96 ± 0.10

6.7

60

1.0 × 1010

8.9

115.55 ± 0.87 5.82 ± 0.08

9.2

60

9

9.1

120.17 ± 2.95 6.45 ± 0.25 21.5

100

4.6 × 108

11.8

104.94 ± 3.14 4.37 ± 0.04 96.3

60

1.5 × 10

Dp: pore diameter, Dint: inter-pore distance, ρ : pore density. E: effective elastic modulus, H: effective hardness. MA: mild anodization, HA: hard anodization. Rq: Root-mean-square surface roughness. Nanoindentation: Berkovich indenter, tip radius is 100–120 nm.

Fig. 1. Schematic diagram of fabricating nanoporous AAO films.

(Dint) were prepared by anodizing aluminum discs using acid electrolytes under regulated cell potentials (U) according either to the conventional mild anodization (MA) or to the newly developed hard anodization (HA) method; sulfuric acid (H2SO4) at 25 VMA for Dint = 63 nm, oxalic acid (H2C2O4) at 40 VMA for Dint = 100 nm and 140 VHA for Dint = 280 nm, and phosphoric acid (H3PO4) at 195 VMA for Dint = 500 nm [10–13]. The surface morphology and the pore

Fig. 3. Friction coefficient of nanoporous structured AAO films with various pore diameters as a function of sliding cycles at 10 mN load.

Fig. 2. SEM images of surfaces of nanoporous structured AAO films.

H. Kim et al. / Surface & Coatings Technology 205 (2010) 1431–1437

Fig. 4. Friction coefficient of nanoporous AAO films at various loads in relation to the pore diameter.

structure of the resulting samples were investigated by scanning electron microscopy (SEM). Pore size (Dp) and pore density (ρ) of the samples were determined by performing image analysis on SEM micrographs. In general, the pore walls of an as-prepared AAO films

1433

are contaminated with acid anions and water originated from the electrolyte, in which the amount of impurities and the chemical structure of pore wall oxide are dependent on the electrochemical conditions employed for anodic oxidation of aluminum [13–15]. Fig. 2 shows scanning electron micrographs of the surfaces of the fabricated specimens. Measurements of hardness and elastic modulus of the AAO films were carried out using a MTS nano-indenter XP. 9 indentations were made for each sample to secure reliability and the values obtained between 100 and 500 nm indentation depth were used to calculate the nominal value. The porosity was measured using an image analyzer. Table 1 shows the effective hardness and modulus values in association with the pore diameter and porosity. The indentation results showed that both the effective hardness and elastic modulus of the films fabricated by mild anodization were greatly influenced by the porosity and the pore size. However, the sample produced by hard anodization had a relatively higher value than those with comparable porosity obtained by mild anodization. Previous study indicates that anodic films form by hard anodization process exhibit higher density (ca. 3.1 g/cm3) compared to those (ca. 2.8 g/cm3) of mild anodized ones [13]. We believe that the higher density of hard anodized AAO film is responsible for the improved hardness. We employed an atomic force microscope to characterize surface topography. As it can be noted from the table, surface roughness values were in the range of 6.7–96.3 nm (Rq); the larger pore size (pore diameter), the rougher

Fig. 5. SEM images of tribolayer formed on the worn surfaces of nanoporous AAO films.

1434

H. Kim et al. / Surface & Coatings Technology 205 (2010) 1431–1437

due to the fact that the scanned area for every measurement contained not only flat region but also pores. The surface roughness of steel balls as the counterparts was about 20 nm (Rq), implying its negligible influence on the tribological behavior of AAO films. Contact pressure and radius of the contact area for each load were calculated based on Hertzian contact theory where pore-free alumina surface were assumed for the calculation. 2.2. Friction and wear test For tribological measurement, we employed a reciprocating wear tester (UMT-2) using two types of load sensor, FVL and DFM-0.5, which allows the application of loads in 4 orders of magnitude range. We used a 1 mm diameter 440C stainless steel ball as a counterpart, whose elastic modulus and Poisson's ratio were 200 MPa and 0.28, respectively, and hardness was HRB 97. The upper ball counterpart reciprocated with a velocity of 0.5 mm/s against AAO specimens with a stroke length of 3 mm. 4 different loads were applied from 1 mN to 1 N with sliding frequency of 100 cycles at ambient environment. The loads 1 mN, 10 mN, 100 mN and 1 N corresponded to the maximum Hertzian pressure of 103.1, 222.1, 478.5 and 1030.5 MPa and the contact radius of 2.2, 4.6, 10.1 and 21.5 μm, respectively. Prior to tests, the ball specimens were subjected to ultrasonic cleaning in ethanol for 15 min each. At least three tests were run for each load to check the reproducibility of the friction behavior. After each test was completed, the worn surfaces of

the AAO and steel ball wear scars were observed using SEM. EDS analysis was performed to examine the elemental characteristics of the wear tracks of the AAO films and the counterpart steel balls. 3. Results and discussion Fig. 3 shows the variation of the friction coefficient as a function of a number of sliding cycles at 10 mN load. As a reference pore-free AAO film was also tested and presented in the figure. Other specimens also exhibited similar behavior to that with 10 mN load. The friction coefficient increased at the initial period of the test (b20 cycles) followed by a decrease to a relatively steady level that was then maintained until the end of the test runs (100 cycles). Fig. 4 illustrates that the steady friction coefficient significantly reduced with the increase of the load for all AAO films with different pore sizes. At the low loads (1 mN and 10 mN), the friction coefficient tended to increase with the increase of the pore size. On the other hand, at relatively high loads (0.1 N and 1 N), the influence of pore size on friction coefficient was negligible. The worn surfaces of AAO films were investigated using SEM. Figs. 5 and 6 showed scanning electron micrographs of worn surfaces of AAO films and wear particles deposited on the wear tracks, respectively. The worn surfaces of AAO films tested at low loads (1 mN and 10 mN) showed an insignificant sliding damage but the worn surface tested at 10 mN revealed sparsely distributed extremely thin plastically

Fig. 6. SEM images of wear particles generated and deposited on the worn surfaces of nanoporous AAO films.

H. Kim et al. / Surface & Coatings Technology 205 (2010) 1431–1437

deformed layers. As the plastically deformed layers were extremely thin, the AAO nano-pores underneath the layers were detectable. The microasperities of the AAO film were abraded and formed micro-particles by sliding contact with the counterpart steel ball, and then they were trapped in the contact interface and eventually formed the extremely thin layers by agglomeration and plastic deformation. On the other hand, severe friction was evident at high loads (0.1 N and 1 N). Fig. 5 demonstrated that thick tribolayers were formed on the worn surface at high loads. Furthermore some cracks were observed along the boundary of the wear track at 1 N load. These thick smooth layers were generated by combined influence of tribochemical reaction including material transfer at the contact interface and severe plastic deformation of compacted debris particles. The wear particles generated at low loads (1 mN and 10 mN) shaped micro-cylinders due to the repeated rolling of micro-wear particles at the contact interface with reciprocated sliding under very low contact pressure (Fig. 6). In contrast, relatively large and thin plate-like wear particles were dominantly generated at high loads as evident in Fig. 6 due to delamination of tribolayers by repeated sliding under high load. The worn surfaces of counterpart steel balls also exhibited similar behaviors to those observed for AAO films (Fig. 7). In this study, we observed that the tribolayer formation increased with the increase of load. The positive contribution of tribolayers by decreasing friction at the contact interface has been widely observed in various ranges of load with various contact pairs [16–18]. By considering the fact that the friction coefficient considerably

1435

decreased with the increase of load for all cases, we confirmed that these tribolayers contributed to the lower friction. The chemical components of tribolayers and wear particles at each load were analyzed using EDS as presented in Figs. 8 and 9. Fe and Cr elements which are main elements of steel ball were detected in the thick tribolayers of wear tracks of AAO films (Fig. 8) while Al and S elements, the main chemical elements of AAO, were detected in the thick tribolayer of steel ball (Fig. 9). It explains that active tribochemical reaction and material transfer occurred between the mating materials. However, as noted with the low loading case in Figs. 8 and 9, the extremely thin layer patches, due to mild plastic deformation of surface layer, did not show any evidence of tribochemical reaction. Therefore, it is believed that active tribochemical reaction and material transfer between AAO and steel ball occurred at only high loads. This phenomenon is surely due to the contribution of higher friction heating at high loads [19]. We confirmed that AAO films themselves do not exhibit satisfactory tribological performance and therefore they should be utilized as a template for forming self-lubricating structures. Although several investigations have been made in this regard as briefly illustrated in the introduction, no attempt has yet been made saturating oil molecules in the pores for use as boundary lubricant. Nanocomposite porous elastomeric polymer layers saturated with paraffinic oil resulted in a significant reduction of the local friction forces [20]. These internally lubricated coatings show a much lower

Fig. 7. SEM images of worn surfaces of the counterpart steel ball slid against AAO film with 95 nm pore diameter.

1436

H. Kim et al. / Surface & Coatings Technology 205 (2010) 1431–1437

Fig. 8. SEM images and EDS analysis results for tribolayers and wear particles deposited on the worn surface of the nanosturctured AAO film with 28 nm pore diameter (the applied load is noted in the inset of SEM image).

friction coefficient and higher life-time under high load than a classic organic boundary lubricant, alkylsilane SAM. The authors suggested that long-chain molecules from the oil phase facilitate the ‘contact’ lubrication that served for the instant modification of the local area of the mechanical contact via the mechanism of contact pressure driven expulsion and surface diffusion of long-chain molecules from inside the rubber matrix. In further research, we focus on saturating oil molecules in the pores and obtain enhanced tribological performance of the nanocoatings via reducing oil phase mobility and facilitating the mechanism of local contact melting for higher performance of nanocomposite coatings.

4. Conclusions We investigated the nanoporous structured AAO films with various pore sizes to better understand their tribological behavior in

sliding contact with steel ball with 4 orders of magnitude range of normal load (from 1 mN to 1 N). The results of the present investigation can be summarized as follows: 1. The influence of pore size (pore diameter) of AAO films on the friction coefficient was dominant at relatively high loads (0.1 N and 1 N): the larger the pore size, the higher the friction coefficient. 2. The friction coefficient considerably reduced with the increase of load for all AAO films independent of their pore sizes. 3. Smooth and thick tribolayer patches were formed on the worn surface of AAO at relatively high loads (0.1 N and 1 N) due to tribochemical reaction and compaction of wear debris. These smooth and thick tribolayer patches contributed to the lower friction and wear at high loads. 4. Extremely thin tribolayer patches, due to mild plastic deformation of surface layer and/or micro-wear particles, were sparsely distributed on the worn surface of AAO at 10 mN load without

H. Kim et al. / Surface & Coatings Technology 205 (2010) 1431–1437

1437

Fig. 9. SEM images and EDS analysis results for tribolayers and wear particles deposited on the worn surface of the counterpart steel ball slid against AAO film with 28 nm pore diameter (the applied load is noted in the inset of SEM image).

the evidence of tribochemical reaction. These layers however also contributed to decrease friction despite of their limited thickness. Acknowledgement This work is supported by Korea Research Council of Public Science and Technology (KORP) through the “Development of Advanced Materials Metrology” project. References [1] [2] [3] [4] [5] [6]

P. Kohli, M. Wirtz, C.R. Martin, Electroanalysis 16 (2004) 9. G.L. Che, B.B. Lakshmi, E.R. Fisher, C.R. Martin, Nature 393 (1998) 346. P. Skeldom, H.W. Wang, G.E. Thompson, Wear 206 (1997) 187. M. Maejimaa, U.K. Saruwataria, M. Takaya, Surf. Coat. Technol. 132 (2000) 105. K. Hiratsuka, M. Asakawa, A. Funakoshi, M. Takaya, Trib. Lett. 13 (2002) 77. M. Takaya, K. Hashimoto, Y. Toda, M. Maejima, Surf. Coat. Technol. 169 (2003) 160.

[7] J.P. Tu, C.X. Jiang, S.Y. Guo, L.P. Zhu, F.M. Fu, X.B. Zha, Surf. Coat. Technol. 198 (2005) 464. [8] J.P. Tu, C.X. Jiang, S.Y. Guo, X.B. Zhao, M.F. Fu, Wear 259 (2005) 759. [9] D. Choi, S. Lee, S. Kim, P. Lee, K. Lee, H.I. Parka, W. Hwang, Scr. Mater. 58 (2008) 870. [10] H. Masuda, M. Satoh, Jpn. J. Appl. Phys. 35 (1996) L126. [11] A.P. Li, F. Muller, A. Birner, K. Nielsch, U. Gösele, J. Appl. Phys. 84 (1998) 6023. [12] W. Lee, R. Scholz, K. Nielsch, U. Gösele, Angew. Chem. Int. Ed. 44 (2005) 6050. [13] W. Lee, R. Ji, U. Gösele, K. Nielsch, Nature Mater. 5 (2006) 741. [14] G.E. Thompson, G.C. Wood, Nature 290 (1981) 230. [15] W. Lee, K. Schwirn, M. Steinhart, E. Pippel, R. Scholz, U. Gösele, Nature Nanotech. 3 (2008) 402. [16] I.-W. Lyo, H.-S. Ahn, D.-S. Lim, Surf. Coat. Technol. 163-164 (2002) 395. [17] H.-S. Ahn, I.-W. Lyo, D.-S. Lim, Surf. Coat. Technol. 133-134 (2000) 351. [18] H.S. Ahn, P.D. Cuong, K.H. Shin, K.S. Lee, Wear 259 (2005) 807. [19] T.E. Fischer, Ann. Rev. Mater. Sci. 18 (1988) 303. [20] D. Julthongpiput, H.-S. Ahn, A. Sidorenko, D.-I. Kim, V.V. Tsukruk, Tribol. Int. 35 (2002) 829.